Novel genomic mechanism for ligand-dependent transcription by androgen receptor Project Summary/Abstract Androgen receptor (AR) is a member of nuclear hormone receptor (NR) superfamily that binds to cognate hormone responsive elements (HREs) and regulates target gene expression in an endogenous ligand (agonist)-inducible manner in diverse tissues. As AR plays a key role in the onset and progression of prostate cancer, numerous synthetic AR antagonists have been developed to inhibit the action of endogenous AR ligands. A prominent example is enzalutamide (Xtandi), a second-generation AR antagonist showing strong anti-cancer activity for prostate cancer. However, intrinsic or acquired resistance to enzalutamide, and all available AR antagonists, occurs leading to treatment failure. Thus, therapeutic efficacy of current AR antagonists needs to be improved. Elucidation of genomic mechanisms underlying antagonist-liganded AR function is critically important in order to improve AR-targeted therapy. In preliminary studies, we have defined the first high-resolution (motif-resolution) agonist- and antagonist-liganded AR cistromes in prostate cancer cells by using a novel chromatin immunoprecipitation-exonuclease (ChIP-exo) approach. Unexpectedly, we found that AR bound to natural agonist (dihydrotestosterone, DHT) and antagonist (enzalutamide) recognizes distinctly different DNA motifs on chromatin (termed ?DNA motif switching?). Surprisingly, integrated ChIP-exo and RNA-seq analysis found that enzalutamide-liganded AR, bound to a novel AR binding motif, significantly affects global, cancer-relevant transcription. By combining our novel ChIP-exo genomic approach with other epigenomic, proteomic and biochemical approaches, we further found that enzalutamide-liganded AR interacts with specific collaborating transcription factors (e.g. FoxA1) and non-DNA binding coregulators (e.g. Hsp90) on specific active cis-regulatory regions. Importantly, pharmacological Hsp90 inhibition significantly decreases expression of enzalutamide-liganded AR target genes (e.g. cancer promoting genes GR and CD55) and enhances cell growth inhibitory effect of enzalutamide. Based on these compelling data, we hypothesize that DNA motif switching is a novel genomic mechanism underlying antagonist-dependent, cancer-relevant transcription by antagonist-liganded AR transcription complex. Our specific aims are to: 1) determine whether specific transcription factors and epigenetic features globally facilitate AR DNA motif switching; and 2) investigate how antagonist-liganded AR binding regulates expression of cancer-relevant genes. By significantly enhancing our understanding of how antagonist-regulated transcription by AR is controlled at the genomic level, this study will lay the foundation for future development of improved AR-targeted therapy.